import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import os
for dirname, _, filenames in os.walk('/kaggle/input'):
    for filename in filenames:
        print(os.path.join(dirname, filename))

import seaborn as sns
import matplotlib.pyplot as plt

df=pd.read_csv('deepseek_vs_chatgpt.csv')

df

df.columns

Index(['Date', 'Month_Num', 'Weekday', 'AI_Platform', 'AI_Model_Version',
       'Active_Users', 'New_Users', 'Churned_Users', 'Daily_Churn_Rate',
       'Retention_Rate', 'User_ID', 'Query_Type', 'Input_Text',
       'Input_Text_Length', 'Response_Tokens', 'Topic_Category', 'User_Rating',
       'User_Experience_Score', 'Session_Duration_sec', 'Device_Type',
       'Language', 'Response_Accuracy', 'Response_Speed_sec',
       'Response_Time_Category', 'Correction_Needed', 'User_Return_Frequency',
       'Customer_Support_Interactions', 'Region'],
      dtype='object')

df.describe()

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 10000 entries, 0 to 9999
Data columns (total 28 columns):
 #   Column                         Non-Null Count  Dtype  
---  ------                         --------------  -----  
 0   Date                           10000 non-null  object 
 1   Month_Num                      10000 non-null  int64  
 2   Weekday                        10000 non-null  object 
 3   AI_Platform                    10000 non-null  object 
 4   AI_Model_Version               10000 non-null  object 
 5   Active_Users                   10000 non-null  int64  
 6   New_Users                      10000 non-null  int64  
 7   Churned_Users                  10000 non-null  int64  
 8   Daily_Churn_Rate               10000 non-null  float64
 9   Retention_Rate                 10000 non-null  float64
 10  User_ID                        10000 non-null  object 
 11  Query_Type                     10000 non-null  object 
 12  Input_Text                     10000 non-null  object 
 13  Input_Text_Length              10000 non-null  int64  
 14  Response_Tokens                10000 non-null  int64  
 15  Topic_Category                 10000 non-null  object 
 16  User_Rating                    10000 non-null  int64  
 17  User_Experience_Score          10000 non-null  float64
 18  Session_Duration_sec           10000 non-null  int64  
 19  Device_Type                    10000 non-null  object 
 20  Language                       10000 non-null  object 
 21  Response_Accuracy              9621 non-null   float64
 22  Response_Speed_sec             10000 non-null  float64
 23  Response_Time_Category         10000 non-null  object 
 24  Correction_Needed              10000 non-null  int64  
 25  User_Return_Frequency          10000 non-null  int64  
 26  Customer_Support_Interactions  10000 non-null  int64  
 27  Region                         10000 non-null  object 
dtypes: float64(5), int64(11), object(12)
memory usage: 2.1+ MB

df.tail(5)

df.isnull().sum()

Date                               0
Month_Num                          0
Weekday                            0
AI_Platform                        0
AI_Model_Version                   0
Active_Users                       0
New_Users                          0
Churned_Users                      0
Daily_Churn_Rate                   0
Retention_Rate                     0
User_ID                            0
Query_Type                         0
Input_Text                         0
Input_Text_Length                  0
Response_Tokens                    0
Topic_Category                     0
User_Rating                        0
User_Experience_Score              0
Session_Duration_sec               0
Device_Type                        0
Language                           0
Response_Accuracy                379
Response_Speed_sec                 0
Response_Time_Category             0
Correction_Needed                  0
User_Return_Frequency              0
Customer_Support_Interactions      0
Region                             0
dtype: int64

df['Response_Accuracy'].fillna(df['Response_Accuracy'].median(), inplace=True)

C:\Users\ABHISHEK\AppData\Local\Temp\ipykernel_21244\505380775.py:1: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.
The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.

For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.


  df['Response_Accuracy'].fillna(df['Response_Accuracy'].median(), inplace=True)

def detect_outliers_iqr(df, column):
    Q1 = df[column].quantile(0.25)
    Q3 = df[column].quantile(0.75)
    IQR = Q3 - Q1
    lower_bound = Q1 - 1.5 * IQR
    upper_bound = Q3 + 1.5 * IQR
    outliers = df[(df[column] < lower_bound) | (df[column] > upper_bound)]
    return outliers

# Columns to check for outliers
num_cols = ['Response_Accuracy', 'User_Rating', 'Session_Duration_sec', 'Response_Speed_sec']

# Detect outliers in selected numerical columns
for col in num_cols:
    outliers = detect_outliers_iqr(df, col)
    print(f"Outliers in {col}: {len(outliers)}")

Outliers in Response_Accuracy: 16
Outliers in User_Rating: 0
Outliers in Session_Duration_sec: 0
Outliers in Response_Speed_sec: 0

def replace_outliers(df, column, method="median"):
    Q1 = df[column].quantile(0.25)
    Q3 = df[column].quantile(0.75)
    IQR = Q3 - Q1
    lower_bound = Q1 - 1.5 * IQR
    upper_bound = Q3 + 1.5 * IQR

    if method == "median":
        replacement = df[column].median()
    else:
        replacement = df[column].mean()
    
    df[column] = np.where((df[column] < lower_bound) | (df[column] > upper_bound), replacement, df[column])

# Apply to numerical columns
for col in num_cols:
    replace_outliers(df, col, method="median")

print("Outliers replaced with median values!")

Outliers replaced with median values!

def remove_outliers(df, column):
    Q1 = df[column].quantile(0.25)
    Q3 = df[column].quantile(0.75)
    IQR = Q3 - Q1
    lower_bound = Q1 - 1.5 * IQR
    upper_bound = Q3 + 1.5 * IQR
    return df[(df[column] >= lower_bound) & (df[column] <= upper_bound)]

# Apply to selected numerical columns
for col in num_cols:
    df = remove_outliers(df, col)

print("Outliers removed!")

Outliers removed!

plt.figure(figsize=(12, 6))
for i, col in enumerate(num_cols, 1):
    plt.subplot(2, 2, i)
    sns.boxplot(x=df[col], color='green')
    plt.title(f"Boxplot After Outlier Handling: {col}")
plt.tight_layout()
plt.show()

plt.figure(figsize=(8, 5))
sns.histplot(df['User_Rating'], bins=10, kde=True, color='blue')
plt.title("Distribution of User Ratings")
plt.xlabel("User Rating")
plt.ylabel("Frequency")
plt.show()

plt.figure(figsize=(8, 5))
sns.boxplot(x=df['Response_Accuracy'], color='green')
plt.title("Boxplot of Response Accuracy")
plt.show()

plt.figure(figsize=(8, 5))
sns.barplot(x=df['AI_Platform'], y=df['Active_Users'], palette="coolwarm")
plt.title("Active Users per AI Platform")
plt.xlabel("AI Platform")
plt.ylabel("Number of Active Users")
plt.show()

C:\Users\ABHISHEK\AppData\Local\Temp\ipykernel_21244\1770559313.py:2: FutureWarning: 

Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.

  sns.barplot(x=df['AI_Platform'], y=df['Active_Users'], palette="coolwarm")

plt.figure(figsize=(8, 5))
sns.scatterplot(x=df['Response_Accuracy'], y=df['User_Rating'], hue=df['AI_Platform'])
plt.title("User Rating vs Response Accuracy")
plt.xlabel("Response Accuracy")
plt.ylabel("User Rating")
plt.show()

plt.figure(figsize=(8, 5))
sns.boxplot(x=df['AI_Platform'], y=df['Response_Accuracy'], palette="Set2")
plt.title("Response Accuracy by AI Platform")
plt.show()

C:\Users\ABHISHEK\AppData\Local\Temp\ipykernel_21244\3714208981.py:2: FutureWarning: 

Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.

  sns.boxplot(x=df['AI_Platform'], y=df['Response_Accuracy'], palette="Set2")

plt.figure(figsize=(10, 5))
sns.countplot(x=df['AI_Model_Version'], hue=df['AI_Platform'], palette="Set1")
plt.xticks(rotation=45)
plt.title("AI Model Version Distribution")
plt.show()

import plotly.express as px
fig = px.scatter(df, x='Response_Speed_sec', y='User_Experience_Score', color='AI_Platform',
                 title="Response Speed vs User Experience Score", size='User_Experience_Score')
fig.show()

fig = px.bar(df, x='Region', y='Active_Users', color='AI_Platform', 
             title="Active Users per Region", barmode="group")
fig.show()

from wordcloud import WordCloud, STOPWORDS
import nltk
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
plt.figure(figsize=(12, 6))

for platform in df['AI_Platform'].unique():
    platform_text = " ".join(df[df['AI_Platform'] == platform]['Input_Text'].dropna())
    platform_wordcloud = WordCloud(width=800, height=400, background_color='black',
                                   colormap="plasma").generate(platform_text)
    
    plt.subplot(1, 2, list(df['AI_Platform'].unique()).index(platform) + 1)
    plt.imshow(platform_wordcloud, interpolation="bilinear")
    plt.axis("off")
    plt.title(f"Most Searched Terms - {platform}")

plt.tight_layout()
plt.show()

def generate_wordcloud(column_name, bg_color="white", cmap="coolwarm"):
    plt.figure(figsize=(10, 5))

    # Combine all text from the column
    text = " ".join(df[column_name].dropna().astype(str))

    # Create WordCloud
    wordcloud = WordCloud(width=800, height=400, background_color=bg_color,
                          stopwords=STOPWORDS, colormap=cmap).generate(text)

    # Display WordCloud
    plt.imshow(wordcloud, interpolation="bilinear")
    plt.axis("off")
    plt.title(f"WordCloud for {column_name}", fontsize=14)
    plt.show()

generate_wordcloud("Query_Type")

generate_wordcloud("Topic_Category", bg_color="black", cmap="plasma")

generate_wordcloud("Language", bg_color="white", cmap="viridis")

generate_wordcloud("Device_Type", bg_color="white", cmap="inferno")

plt.figure(figsize=(12, 6))
sns.lineplot(data=df, x="Date", y="Active_Users", hue="AI_Platform", marker="o", linewidth=2)
plt.xticks(rotation=45)
plt.title("Active Users Trend Over Time (Deepseek vs ChatGPT)", fontsize=14)
plt.xlabel("Date")
plt.ylabel("Active Users")
plt.legend(title="AI Platform")
plt.grid(True)
plt.show()

fig = px.bar(df, x="AI_Model_Version", y="Active_Users", color="AI_Platform",
             title="AI Model Version Usage Over Time", barmode="stack")
fig.show()

sns.jointplot(data=df, x="Response_Accuracy", y="User_Rating", hue="AI_Platform", kind="scatter", height=7)
plt.suptitle("User Rating vs Response Accuracy", fontsize=14)
plt.show()

plt.figure(figsize=(10, 5))
sns.heatmap(df.head(25).pivot_table(index="Session_Duration_sec", columns="User_Experience_Score", 
                           values="Active_Users", aggfunc="sum").fillna(0), cmap="coolwarm", annot=True)
plt.title("User Engagement Heatmap: Session Duration vs Experience Score")
plt.xlabel("User Experience Score")
plt.ylabel("Session Duration (sec)")
plt.show()

fig = px.choropleth(df, locations="Region", locationmode="country names", 
                     color="Active_Users", hover_name="Region",
                     title="Most Active Regions (Deepseek vs ChatGPT)", color_continuous_scale="viridis")
fig.show()

fig = px.scatter(df, x="Response_Speed_sec", y="User_Experience_Score", 
                 size="User_Experience_Score", color="AI_Platform",
                 hover_data=["AI_Model_Version", "Response_Accuracy"],
                 title="Response Speed vs User Experience Score")
fig.show()

fig = px.sunburst(df, path=["AI_Platform", "Topic_Category", "Query_Type"], 
                  values="Active_Users", color="AI_Platform",
                  title="AI Query Distribution Across Topics")
fig.show()

fig, ax1 = plt.subplots(figsize=(12, 6))

# First line plot for Daily Churn Rate
sns.lineplot(data=df, x="Date", y="Daily_Churn_Rate", hue="AI_Platform", marker="o", ax=ax1)
ax1.set_ylabel("Daily Churn Rate (%)", color="red")
ax1.tick_params(axis="y", labelcolor="red")

# Second line plot for Retention Rate on the same axis
ax2 = ax1.twinx()
sns.lineplot(data=df, x="Date", y="Retention_Rate", hue="AI_Platform", marker="s", linestyle="dashed", ax=ax2)
ax2.set_ylabel("Retention Rate (%)", color="blue")
ax2.tick_params(axis="y", labelcolor="blue")

plt.title("Daily Churn Rate vs Retention Rate Over Time", fontsize=14)
plt.xticks(rotation=45)
plt.show()

plt.figure(figsize=(12, 6))
sns.countplot(data=df, y="Query_Type", hue="AI_Platform", order=df["Query_Type"].value_counts().index, palette="viridis")
plt.title("Most Common Query Types by AI Platform", fontsize=14)
plt.xlabel("Count")
plt.ylabel("Query Type")
plt.show()

fig = px.scatter(df, x="Response_Accuracy", y="User_Experience_Score", size="Active_Users", color="Topic_Category",
                 hover_data=["AI_Platform"], title="Response Accuracy vs User Experience by Topic")
fig.show()

df_device = df["Device_Type"].value_counts().reset_index()
df_device.columns = ["Device_Type", "Count"]

plt.figure(figsize=(8, 8))
plt.pie(df_device["Count"], labels=df_device["Device_Type"], autopct="%1.1f%%", startangle=90, colors=["blue", "green", "orange", "red"])
plt.title("Device Type Distribution")
plt.show()

plt.figure(figsize=(10, 5))
sns.boxplot(data=df, x="AI_Platform", y="Customer_Support_Interactions", palette="coolwarm")
plt.title("Customer Support Interactions Across AI Platforms", fontsize=14)
plt.xlabel("AI Platform")
plt.ylabel("Support Interactions")
plt.show()

C:\Users\ABHISHEK\AppData\Local\Temp\ipykernel_21244\3627237335.py:2: FutureWarning:


Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.

plt.figure(figsize=(10, 6))
heatmap_data = df.head(150).pivot_table(index="Weekday", columns="Month_Num", values="Active_Users", aggfunc="sum")
sns.heatmap(heatmap_data, cmap="coolwarm", annot=True, fmt=".0f")
plt.title("Active Users Heatmap: Weekdays vs Months", fontsize=14)
plt.xlabel("Month")
plt.ylabel("Weekday")
plt.show()

plt.figure(figsize=(10, 6))
sns.violinplot(data=df, x="AI_Platform", y="User_Return_Frequency", palette="pastel", inner="quartile")
plt.title("User Return Frequency Across AI Platforms", fontsize=14)
plt.xlabel("AI Platform")
plt.ylabel("User Return Frequency")
plt.show()

C:\Users\ABHISHEK\AppData\Local\Temp\ipykernel_21244\4199882007.py:2: FutureWarning:


Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.

import plotly.graph_objects as go

# Ensure 'Date' is in datetime format
df["Date"] = pd.to_datetime(df["Date"])

# Check for AI Platform values
deepseek_data = df[df["AI_Platform"].str.strip().eq("Deepseek")]
chatgpt_data = df[df["AI_Platform"].str.strip().eq("ChatGPT")]

fig = go.Figure()

# Add Deepseek data if available
if not deepseek_data.empty:
    fig.add_trace(go.Scatter(x=deepseek_data["Date"], 
                             y=deepseek_data["Active_Users"], 
                             mode="lines+markers", name="Deepseek", line=dict(color="green")))

# Add ChatGPT data if available
if not chatgpt_data.empty:
    fig.add_trace(go.Scatter(x=chatgpt_data["Date"], 
                             y=chatgpt_data["Active_Users"], 
                             mode="lines+markers", name="ChatGPT", line=dict(color="red")))

fig.update_layout(title="Active Users Over Time (Deepseek vs ChatGPT)",
                  xaxis_title="Date",
                  yaxis_title="Active Users",
                  template="plotly_dark")

fig.show()

import streamlit as st
import pandas as pd
import plotly.express as px
from datetime import datetime
import os

# **Session Tracking File**
SESSION_FILE = "user_sessions.csv"

# **Load Dataset**
@st.cache_data
def load_data():
    return pd.read_csv("deepseek_vs_chatgpt.csv")  # Replace with actual dataset

df = load_data()

# **Function to Load User Sessions**
def load_session_data():
    if os.path.exists(SESSION_FILE):
        return pd.read_csv(SESSION_FILE)
    else:
        return pd.DataFrame(columns=["timestamp", "username", "search", "platform"])

# **Function to Save Session Data**
def save_session(username, search, platform):
    session_data = load_session_data()
    new_entry = pd.DataFrame([{"timestamp": datetime.now(), "username": username, "search": search, "platform": platform}])
    session_data = pd.concat([session_data, new_entry], ignore_index=True)
    session_data.to_csv(SESSION_FILE, index=False)

# **Sidebar Navigation**
st.sidebar.title("Navigation")
page = st.sidebar.radio("Go to", ["📊 Admin Dashboard", "🔍 AI Search Dashboard"])

# **Admin Dashboard**
if page == "📊 Admin Dashboard":
    st.title("📊 Admin Dashboard - User Tracking")

    # **Load User Sessions**
    session_data = load_session_data()

    # **Total Logins**
    st.subheader("👤 Total Logins")
    total_logins = session_data["username"].nunique()
    st.metric(label="Total Users Logged In", value=total_logins)

    # **Most Active Users**
    st.subheader("🔥 Most Active Users")
    user_counts = session_data["username"].value_counts().reset_index()
    user_counts.columns = ["User", "Login Count"]
    st.dataframe(user_counts)

    # **Recent Sessions**
    st.subheader("🕒 Recent User Sessions")
    st.dataframe(session_data.tail(10))

    # **User Search & Filtering**
    st.subheader("🔍 Search User Activity")
    search_user = st.text_input("Enter username to filter activity")
    if search_user:
        user_activity = session_data[session_data["username"] == search_user]
        st.dataframe(user_activity)

# **AI Search Dashboard**
else:
    st.title("🔍 AI Search Dashboard")

    st.sidebar.title("🔍 Filter Options")
    selected_platform = st.sidebar.selectbox("Select AI Platform", df["AI_Platform"].unique())

    search_query = st.sidebar.text_input("Search Query Type")
    username = st.sidebar.text_input("Enter your username")  # Manual user input

    if username:
        save_session(username, search_query, selected_platform)

    filtered_df = df[df["AI_Platform"] == selected_platform]
    if search_query:
        filtered_df = filtered_df[filtered_df["Query_Type"].astype(str).str.contains(search_query, case=False, na=False)]

    st.subheader("📋 Filtered Data")
    st.dataframe(filtered_df)

2025-03-15 20:27:15.117 WARNING streamlit.runtime.caching.cache_data_api: No runtime found, using MemoryCacheStorageManager
2025-03-15 20:27:15.120 WARNING streamlit.runtime.caching.cache_data_api: No runtime found, using MemoryCacheStorageManager
2025-03-15 20:27:15.122 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.355 
  Warning: to view this Streamlit app on a browser, run it with the following
  command:

    streamlit run C:\Users\ABHISHEK\anaconda3\Lib\site-packages\ipykernel_launcher.py [ARGUMENTS]
2025-03-15 20:27:15.356 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.359 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.442 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.443 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.447 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.450 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.452 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.454 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.458 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.461 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.463 Session state does not function when running a script without `streamlit run`
2025-03-15 20:27:15.465 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.467 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.476 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.479 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.483 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.497 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.499 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.500 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.502 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.503 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.534 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.535 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.537 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.538 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.540 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.542 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.544 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.545 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.546 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.547 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.547 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.549 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.549 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2025-03-15 20:27:15.550 Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.

	Date	Month_Num	Weekday	AI_Platform	AI_Model_Version	Active_Users	New_Users	Churned_Users	Daily_Churn_Rate	Retention_Rate	...	Session_Duration_sec	Device_Type	Language	Response_Accuracy	Response_Speed_sec	Response_Time_Category	Correction_Needed	User_Return_Frequency	Customer_Support_Interactions	Region
0	2024-09-21	9	Saturday	ChatGPT	GPT-4-turbo	500000	25000	25000	0.05	0.95	...	40	Mobile	es	0.7842	3.30	Standard	0	6	2	Antarctica (the territory South of 60 deg S)
1	2024-09-21	9	Saturday	ChatGPT	GPT-4-turbo	500000	25000	25000	0.05	0.95	...	24	Laptop/Desktop	zh	0.8194	3.28	Standard	1	2	2	Ukraine
2	2024-09-21	9	Saturday	ChatGPT	GPT-4-turbo	500000	25000	25000	0.05	0.95	...	34	Mobile	en	0.8090	3.07	Standard	0	2	0	Grenada
3	2024-09-21	9	Saturday	ChatGPT	GPT-4-turbo	500000	25000	25000	0.05	0.95	...	18	Mobile	fr	0.8233	3.06	Standard	0	9	0	Guyana
4	2024-05-16	5	Thursday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	10	Mobile	de	0.9366	1.48	Fast	0	9	3	India
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
9995	2024-05-17	5	Friday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	34	Laptop/Desktop	zh	0.9791	0.60	Instant	0	7	2	Seychelles
9996	2024-05-17	5	Friday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	19	Laptop/Desktop	en	0.9132	0.83	Instant	0	5	0	Christmas Island
9997	2024-05-17	5	Friday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	29	Laptop/Desktop	de	0.9516	0.94	Instant	0	10	2	Ethiopia
9998	2024-05-17	5	Friday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	21	Mobile	de	0.9359	0.83	Instant	0	5	3	Puerto Rico
9999	2024-05-17	5	Friday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	58	Mobile	fr	0.9399	0.76	Instant	1	7	1	Kyrgyz Republic

	Month_Num	Active_Users	New_Users	Churned_Users	Daily_Churn_Rate	Retention_Rate	Input_Text_Length	Response_Tokens	User_Rating	User_Experience_Score	Session_Duration_sec	Response_Accuracy	Response_Speed_sec	Correction_Needed	User_Return_Frequency	Customer_Support_Interactions
count	10000.000000	1.000000e+04	10000.000000	10000.000000	10000.000000	1.000000e+04	10000.000000	10000.000000	10000.000000	10000.000000	10000.000000	9621.000000	10000.000000	10000.000000	10000.000000	10000.000000
mean	7.128900	1.196255e+06	100508.750000	35395.150000	0.035228	9.500000e-01	6.260700	274.765100	4.394700	1.626706	28.533700	0.850287	2.356651	0.144600	5.530600	1.476800
std	3.559712	7.444465e+05	85584.077151	14849.189585	0.014999	1.054765e-14	1.188561	130.077225	0.734551	0.491296	14.090348	0.072755	1.303743	0.351715	2.867906	1.120887
min	1.000000	2.000000e+05	12500.000000	4000.000000	0.020000	9.500000e-01	4.000000	50.000000	3.000000	0.480000	5.000000	0.654200	0.330000	0.000000	1.000000	0.000000
25%	4.000000	6.500000e+05	35000.000000	25000.000000	0.020000	9.500000e-01	6.000000	162.000000	4.000000	1.230000	17.000000	0.801800	1.250000	0.000000	3.000000	0.000000
50%	8.000000	9.500000e+05	52500.000000	35000.000000	0.050000	9.500000e-01	7.000000	276.000000	5.000000	1.710000	27.000000	0.862200	2.070000	0.000000	6.000000	1.000000
75%	10.000000	1.700000e+06	170000.000000	49000.000000	0.050000	9.500000e-01	7.000000	386.250000	5.000000	2.070000	38.000000	0.905000	3.450000	0.000000	8.000000	2.000000
max	12.000000	3.050000e+06	305000.000000	61000.000000	0.050000	9.500000e-01	8.000000	500.000000	5.000000	2.280000	60.000000	0.997200	5.190000	1.000000	10.000000	3.000000

	Date	Month_Num	Weekday	AI_Platform	AI_Model_Version	Active_Users	New_Users	Churned_Users	Daily_Churn_Rate	Retention_Rate	...	Session_Duration_sec	Device_Type	Language	Response_Accuracy	Response_Speed_sec	Response_Time_Category	Correction_Needed	User_Return_Frequency	Customer_Support_Interactions	Region
9995	2024-05-17	5	Friday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	34	Laptop/Desktop	zh	0.9791	0.60	Instant	0	7	2	Seychelles
9996	2024-05-17	5	Friday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	19	Laptop/Desktop	en	0.9132	0.83	Instant	0	5	0	Christmas Island
9997	2024-05-17	5	Friday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	29	Laptop/Desktop	de	0.9516	0.94	Instant	0	10	2	Ethiopia
9998	2024-05-17	5	Friday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	21	Mobile	de	0.9359	0.83	Instant	0	5	3	Puerto Rico
9999	2024-05-17	5	Friday	DeepSeek	DeepSeek-Chat 1.5	1700000	170000	34000	0.02	0.95	...	58	Mobile	fr	0.9399	0.76	Instant	1	7	1	Kyrgyz Republic

🎭 AI Battle: Deepseek vs ChatGPT - A Comprehensive Analysis¶

🔍 Which AI Performs Better? Let’s Uncover the Truth!¶

📖 The Story Behind This Analysis¶

Which AI truly delivers the best experience?"¶

📊 What This Notebook Covers¶

🎬 Let the AI Battle Begin! 🔥¶